The purpose of this study is to combine information from variables measuring psychophysiological endopheno-types of smooth pursuit eye movement (SPEM), and the P50 evoked auditory response to investigate the genetics of schizophrenia. Studies will be conducted in two unique populations, which will allow us to an underlying quantitative liability to schizophrenia: 1) 14 moderately-sized Utah pedigrees genotyped with over 500 highly informative microsatellite markers, so that a disease gene will map 5 cM or less from a genetic marker. These families have also been phenotyped with P50, and new pilot SPEM data are promising. 2) A sample from an isolated island population in Palau, Micronesia which will include information from all living schizophrenics and their families. Because it is a population isolate, a gene contributing to disease suscept-ibility may be more likely to reside on a single ancestral haplotype. Experiment 1) Variable Selection: Utah Sample. Variables from P50 and SPEM will be examined in the Utah families to determine which abnormalities co-occur with illness, show substantial variability, show abnormalities in unaffected family members, and finally show overlap across phenotypic domains. Experiment 2) Segregation Analysis: Utah Sample. Using the Utah families, the best quantitative measures from experiment 1 will be tested to determine which are the most consistent with sample patterns of transmission. A composite variable including schizophrenia and the endophenotypes will be generated which will represent a quantitative liability for schizophrenia. Experiment 3) Linkage analysis: Utah Sample. Linkage analyses will be performed using the Utah families described above with the quantitative endophenotypes and liability measure. Experiment 4) Normative studies of the endophenotypes in Palau. A sample of 70 Palauan schizophrenics and 70 matched normal controls will be analyzed to determine the phenotypic distribution of SPEM and P50 variables in Palauan subjects. Experiment 5) Application of Experiments 1-3 in the Palau Sample. Analytic strategies developed and tested in the Utah sample in experiments 1-3 will be applied to the Palau sample. Although regions of potential linkage will take priority in the linkage phase of the Palau sample, a genome scan will also be employed as genes for schizophrenia may differ in this ethnic isolate. Experiment 6) Linkage disequilibrium. In the event that a region of linkage is found, they will test for linkage disequilibrium using techniques that have been applied in similar Finnish populations.